For those on-line databases updated based upon the unit of line or row, if a row of data incurs a slight amount of change at its respective columns, the entire row of data will be updated upon data updating. It generates a vast impact on the databases, which have a large number of rows of data and require updates of many of its rows within a short period of time. E-commerce customer service related on-line databases are such type of databases as they tend to need updating almost every instant, with a row worth of data containing several megabyte or tens of megabytes of data. Consequently, such data updating is both time consuming and resource wasting. Furthermore, upon unsuccessful data importation into databases, splitting of groups will be resorted to, which also leads to the waste of effort and resources.